[ECOS] HAL macros and RTC

[ECOS] HAL macros and RTC

[Fabrice Gautier]

Hi,

In most hal, CYGNUM_HAL_INTERRUPT_xxx macros are used to describe all the
different Interrupts vectors. Most of the CYGNUM_HAL_INTERRUPT_xxx are arch
specific however. For example some arch define CYGNUM_HAL_INTERRUPT_TIMER0
and _TIMER1 whereas other could define _TIMER1 and TIMER2 or just _TIMER if
they have only one. So i assumed that those macros where completely arch
specific and I defined a CYGNUM_HAL_RTC to the interrupt vector of an
hardware clock.

Unfortunately this macros is used as by the kernel (in clock.cxx) as the
interrupt driving the kernel internal clock. So i actually have to define
CYGNUL_HAL_INTERRUPT_RTC as CYGNUM_HAL_INTERRUPT_TIMER and find another name
for ny hardware clock.

Would you think it would be a better idea to let the
CYGNUM_HAL_INTERRUPT_xxx macro arch-specifique and come up with a generic
name to be used by the kernel ? 

The name doesnt matter really, it's just that I got confused on that...

Thanks,

-- 
Fabrice Gautier, 
Fabrice_Gautier@sdesigns.com

-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss

Re: [ECOS] HAL macros and RTC

[Gary Thomas]

On Mon, 2003-01-27 at 17:04, Fabrice Gautier wrote:
> Hi,
> 
> In most hal, CYGNUM_HAL_INTERRUPT_xxx macros are used to describe all the
> different Interrupts vectors. Most of the CYGNUM_HAL_INTERRUPT_xxx are arch
> specific however. For example some arch define CYGNUM_HAL_INTERRUPT_TIMER0
> and _TIMER1 whereas other could define _TIMER1 and TIMER2 or just _TIMER if
> they have only one. So i assumed that those macros where completely arch
> specific and I defined a CYGNUM_HAL_RTC to the interrupt vector of an
> hardware clock.
> 
> Unfortunately this macros is used as by the kernel (in clock.cxx) as the
> interrupt driving the kernel internal clock. So i actually have to define
> CYGNUL_HAL_INTERRUPT_RTC as CYGNUM_HAL_INTERRUPT_TIMER and find another name
> for ny hardware clock.
> 
> Would you think it would be a better idea to let the
> CYGNUM_HAL_INTERRUPT_xxx macro arch-specifique and come up with a generic
> name to be used by the kernel ? 
> 
> The name doesnt matter really, it's just that I got confused on that...
> 

The platforms define CYGNUM_HAL_INTERRUPT_xxx and they are very
much platform specific.  CYGNUM_HAL_INTERRUPT_RTC is the one 
interrupt which must be defined by a platform for the heartbeat
clock/timer, thus the platform neutral name.

Is this not how you expected it to work?

-- 
------------------------------------------------------------
Gary Thomas                 |
MLB Associates              |  Consulting for the
+1 (970) 229-1963           |    Embedded world
http://www.mlbassoc.com/    |
email: <gary@mlbassoc.com>  |
gpg: http://www.chez-thomas.org/gary/gpg_key.asc
------------------------------------------------------------


-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss

Re: [ECOS] HAL macros and RTC

[Nick Garnett]

Fabrice Gautier <Fabrice_Gautier@sdesigns.com> writes:

> 
> I dont know if the "on core timer" is always plugged into the same IRQ line
> for MIPS or PPC,

It is, it's part of the architecture definition.

> but thats not the case for me. I think we should have in
> the guide: 
> 
> "You only have to calculate and enter the proper CYGNUM_HAL_RTC_CONSTANTS
> definitions in the platform CDL file and define CYGNUM_HAL_INTERRUPT_RTC in
> [some include file]"

Good point, I'll take a look at doing that.

> 
> By the way shouldnt the interrupt be a CDL option like the other RTC related
> constants ?

Not really. The interrupt used is fixed by the hardware of the timer,
usually. If a particular platform provided several timers, and if the
HAL implementor chose to allow a choice, then indeed this value might
need to change along with the definitions of the HAL_CLOCK_*()
macros. 

-- 
Nick Garnett - eCos Kernel Architect
http://www.eCosCentric.com/


-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss

RE: [ECOS] HAL macros and RTC

[Fabrice Gautier]

> -----Original Message-----
> From: Jonathan Larmour [mailto:jifl@eCosCentric.com]
> Sent: Monday, January 27, 2003 8:25 PM
> To: Fabrice Gautier
> Cc: 'Gary Thomas'; Ecos-List (E-mail)
> Subject: Re: [ECOS] HAL macros and RTC
> 
> 
> 
> > When I started my port, I listed my interrupt vectors and 
> > called them CYGNUM_HAL_INTERRUPT_xxx, and one of them is
> > for an hardware Clock so i just
> > named it CYGNUM_HAL_INTERRUPT_RTC. And then everything 
> > compiled fine and redboot runs, i didnt realized that 
> > CYGNUM_HAL_INTERRUPT_RTC was more special than the other
> > CYGNUM_HAL_INTERRUPT_xxx until i tried to do
> > multitasking.  
> 
> Read the porting documentation then ;-).

Actually I did it, (after i found the problem), and it says...

"3. Real-time clock interrupts drive the eCos scheduler clock. Many embedded
CPUs have an on-core timer (e.g. SH) or decrementer (e.g. MIPS, PPC) that
can be used, and in this case it will already be supported by the
architecture/variant HAL. You only have to calculate and enter the proper
CYGNUM_HAL_RTC_CONSTANTS definitions in the platform CDL file."
and 
"4. Interrupt decoding usually differs between platforms because the amount
and type of devices on the board differ. In plf_intr.h you must either
extend or replace the vector definitions otherwise provided by the
architecture or variant interrupt headers. You may also have to define
HAL_INTERRUPT_xxx control macros."

I dont know if the "on core timer" is always plugged into the same IRQ line
for MIPS or PPC, but thats not the case for me. I think we should have in
the guide: 

"You only have to calculate and enter the proper CYGNUM_HAL_RTC_CONSTANTS
definitions in the platform CDL file and define CYGNUM_HAL_INTERRUPT_RTC in
[some include file]"

By the way shouldnt the interrupt be a CDL option like the other RTC related
constants ?

Thanks,

-- 
Fabrice Gautier, 
Fabrice_Gautier@sdesigns.com

-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss

Re: [ECOS] HAL macros and RTC

[Jonathan Larmour]

Fabrice Gautier wrote:
>>-----Original Message-----
>>From: Gary Thomas [mailto:gary@mlbassoc.com]
>>Subject: Re: [ECOS] HAL macros and RTC
>>
>>
>>The platforms define CYGNUM_HAL_INTERRUPT_xxx and they are very
>>much platform specific.  CYGNUM_HAL_INTERRUPT_RTC is the one 
>>interrupt which must be defined by a platform for the heartbeat
>>clock/timer, thus the platform neutral name.
>>
>>Is this not how you expected it to work?
> 
> 
> Well, I was just suggesting to change the name of the macros, because i got
> confused.
> What make  CYGNUM_HAL_INTERRUPT_RTC more platform neutral than say
> CYGNUM_HAL_INTERRUPT_UART0 and the others ? 

Just think of it as an exported API of the HAL.

> When I started my port, I listed my interrupt vectors and called them
> CYGNUM_HAL_INTERRUPT_xxx, and one of them is for an hardware Clock so i just
> named it CYGNUM_HAL_INTERRUPT_RTC. And then everything compiled fine and
> redboot runs, i didnt realized that CYGNUM_HAL_INTERRUPT_RTC was more
> special than the other CYGNUM_HAL_INTERRUPT_xxx until i tried to do
> multitasking.  

Read the porting documentation then ;-).

Jifl
-- 
eCosCentric       http://www.eCosCentric.com/       <info@eCosCentric.com>
--[ "You can complain because roses have thorns, or you ]--
--[  can rejoice because thorns have roses." -Lincoln   ]-- Opinions==mine


-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss

RE: [ECOS] HAL macros and RTC

[Fabrice Gautier]

> -----Original Message-----
> From: Gary Thomas [mailto:gary@mlbassoc.com]
> Subject: Re: [ECOS] HAL macros and RTC
> 
> 
> The platforms define CYGNUM_HAL_INTERRUPT_xxx and they are very
> much platform specific.  CYGNUM_HAL_INTERRUPT_RTC is the one 
> interrupt which must be defined by a platform for the heartbeat
> clock/timer, thus the platform neutral name.
> 
> Is this not how you expected it to work?

Well, I was just suggesting to change the name of the macros, because i got
confused.
What make  CYGNUM_HAL_INTERRUPT_RTC more platform neutral than say
CYGNUM_HAL_INTERRUPT_UART0 and the others ? 

When I started my port, I listed my interrupt vectors and called them
CYGNUM_HAL_INTERRUPT_xxx, and one of them is for an hardware Clock so i just
named it CYGNUM_HAL_INTERRUPT_RTC. And then everything compiled fine and
redboot runs, i didnt realized that CYGNUM_HAL_INTERRUPT_RTC was more
special than the other CYGNUM_HAL_INTERRUPT_xxx until i tried to do
multitasking.  

So thats why i'm suggestion another name like CYGNUM_KERNEL_RTC_VECTOR (or
whatever is judged appropriate) maybe change the other names to
CYGNUM_HAL_myarch_INTERRUPT_xxx (but they are many more)... 

But It's not that important...

-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss